Please use this identifier to cite or link to this item: http://scholarbank.nus.edu.sg/handle/10635/48619
Title: FROM SEMANTIC TO EMOTIONAL SPACE IN SENSE SENTIMENT ANALYSIS
Authors: MITRA MOHTARAMI
Keywords: Sentiment Analysis, Emotion Analysis, Sense Sentiment Similarity, Indirect yes/no Question Answer Pair Inference, Sentiment Orientation Prediction
Issue Date: 19-Jun-2013
Source: MITRA MOHTARAMI (2013-06-19). FROM SEMANTIC TO EMOTIONAL SPACE IN SENSE SENTIMENT ANALYSIS. ScholarBank@NUS Repository.
Abstract: This thesis focuses on inferring Sense Sentiment Similarity (SSS) and indicating its effectiveness in Natural Language Processing. SSS models the relevance of words regarding their senses and underlying sentiments. We first investigate the need for developing sentiment similarity measures to accurately capture SSS, rather than using semantic similarity approaches. We then propose emotion-based sentiment similarity measures in which the word senses are modeled as emotional vectors employing the combination of their semantic and emotional spaces. To map the word senses into emotional vectors, we first employ a fixed set of basic human emotions. Then, assuming that the emotions are hidden, we propose hidden emotional models that automatically mine the number and types of hidden emotions in textual contents. We utilize hidden emotional vectors to predict SSS between word senses. Extensive experiments on Indirect yes/no Question Answer Pair inference and Sentiment Orientation prediction tasks show the effectiveness of the proposed approaches.
URI: http://scholarbank.nus.edu.sg/handle/10635/48619
Appears in Collections:Ph.D Theses (Open)

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
MohtaramiMitra.pdf1.87 MBAdobe PDF

OPEN

NoneView/Download

Page view(s)

130
checked on Dec 11, 2017

Download(s)

30
checked on Dec 11, 2017

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.